what specifically is taking a long time ?
building the remediation objects ?
are there about the same number of targets in the smart group vs the static group ?
The number of servers per SG is the same, the only different is they are smart groups now. The reason I posed the question, is my management want's to rule out the possibility of Smart Groups being the root cause. In this particular instance we were targeting about 200 server per smart group, and sending to 4 smart groups somewhat in parallel. Separate team members managing SOPs etc...
My other question is, can the fact these smart groups where created using BLCLI (a week before) be causing this problem? I don't think this is at all related, but management wanted me to ask.
We did notice there was an issue with one of the patch downloads, we pulled it from the catalog but it didn't seem to make a difference.
can you elaborate on what part of the patching process is slow ? it's hard to say what the problem is until we understand that.
Bill - now that you've seen our environment, do you think Smart Groups are the cause of our performance issues? I'd like to keep moving forward with the new Smart Group strategy, but would need to be certain this is not the issue. I don't think this is root cause, (rather coincidence). As a test, I ran the same job against a Smart Group, and a Static Group (with the Smart Group members copied in), and got the same exact results.
Also, are there any plans to integrate more with Microsoft PowerShell (maybe a BladeLogic PowerShell toolkit)? I'm integrating several processes across heterogeneous systems, and PowerShell is the lowest common denominator. It would be awesome to leverage a BladeLogic toolkit vs. running embedded BLCLI commands in PowerShell.
Lastly, is there a was to reuse the BLCLI JVM in PowerShell, marshal data between the runtimes vs. spawning a new JVM with each BLCLI call?
I don't think there's a problem w/ the smart groups - i mean, there could be. if you have a huge list of conditions in the smart group definition, then yeah, that could take a while to resolve. i'm not sure what 'huge' would be - 50, 100. ~10 shouldn't be a problem.
from the call yesterday the only thing that seemed slow for sure was the gui while the job load was high. it's possible the users are mistaking the slow response time in the gui for slow job run time and that may not be the case.
there's no way that i know to use the blcli jvm in powershell... you can use soap and rest to interact w/ the appserver and powershell should have modules for those.
Thanks Bill, yeah as far as the smart groups they each have between 5 and 7 conditions, nothing major. Can you point me towards PowerShell modules/sample code/anything PowerShell to interact with the SOAP API?
Thanks in advance